from urllib.request import urlopen
import json
with urlopen('https://raw.githubusercontent.com/PublicaMundi/MappingAPI/master/data/geojson/us-states.json') as response:
    states = json.load(response)
    
import pandas as pd
df = pd.read_csv("https://raw.githubusercontent.com/visualization-project-group/surprise-map-assignment/master/hate_crime.csv",usecols=['STATE_NAME', 'BIAS_DESC', 'DATA_YEAR'])
df2 = df.groupby(['STATE_NAME','DATA_YEAR']).agg('count')
df2[:50]

		BIAS_DESC
STATE_NAME	DATA_YEAR
Alabama	1992	4
	1993	5
	2002	2
	2003	1
	2004	3
	2007	7
	2008	11
	2009	9
	2010	19
	2011	83
	2012	6
	2013	6
	2014	9
	2015	10
	2016	14
	2017	14
	2018	2
	2019	5
Alaska	1993	24
	1994	9
	1995	8
	1996	9
	1997	10
	1999	5
	2000	4
	2001	20
	2002	7
	2003	13
	2004	9
	2005	4
	2006	6
	2007	8
	2008	8
	2009	9
	2010	7
	2011	8
	2012	6
	2013	8
	2014	6
	2015	8
	2016	11
	2017	4
	2018	5
	2019	15
Arizona	1991	48
	1992	172
	1993	208
	1994	205
	1995	220
	1996	250

Caculate Anti Asian Rate over all kinds of hate crimes for each year:

df2['Anti-Asian-count'] = df.query("BIAS_DESC == 'Anti-Asian'").groupby(['STATE_NAME', 'DATA_YEAR']).agg('count')
df2 = df2.fillna(0)
df2['Anti-Asian-rate'] =  df2['Anti-Asian-count']/df2['BIAS_DESC']
df2[:50]

		BIAS_DESC	Anti-Asian-count	Anti-Asian-rate
STATE_NAME	DATA_YEAR
Alabama	1992	4	0.0	0.000000
	1993	5	0.0	0.000000
	2002	2	0.0	0.000000
	2003	1	0.0	0.000000
	2004	3	0.0	0.000000
	2007	7	1.0	0.142857
	2008	11	0.0	0.000000
	2009	9	0.0	0.000000
	2010	19	0.0	0.000000
	2011	83	1.0	0.012048
	2012	6	0.0	0.000000
	2013	6	0.0	0.000000
	2014	9	1.0	0.111111
	2015	10	0.0	0.000000
	2016	14	0.0	0.000000
	2017	14	0.0	0.000000
	2018	2	0.0	0.000000
	2019	5	0.0	0.000000
Alaska	1993	24	2.0	0.083333
	1994	9	0.0	0.000000
	1995	8	1.0	0.125000
	1996	9	0.0	0.000000
	1997	10	1.0	0.100000
	1999	5	0.0	0.000000
	2000	4	0.0	0.000000
	2001	20	0.0	0.000000
	2002	7	0.0	0.000000
	2003	13	1.0	0.076923
	2004	9	1.0	0.111111
	2005	4	0.0	0.000000
	2006	6	0.0	0.000000
	2007	8	0.0	0.000000
	2008	8	1.0	0.125000
	2009	9	0.0	0.000000
	2010	7	0.0	0.000000
	2011	8	0.0	0.000000
	2012	6	1.0	0.166667
	2013	8	0.0	0.000000
	2014	6	0.0	0.000000
	2015	8	0.0	0.000000
	2016	11	1.0	0.090909
	2017	4	0.0	0.000000
	2018	5	0.0	0.000000
	2019	15	0.0	0.000000
Arizona	1991	48	2.0	0.041667
	1992	172	8.0	0.046512
	1993	208	9.0	0.043269
	1994	205	6.0	0.029268
	1995	220	9.0	0.040909
	1996	250	5.0	0.020000

Select the mean of 2009-2018 Anti Asian Rate in each state as our Model:

df3 = df2.query('DATA_YEAR>=2009 & DATA_YEAR<=2018').groupby(['STATE_NAME']).agg('mean')
df3[:50]

	BIAS_DESC	Anti-Asian-count	Anti-Asian-rate
STATE_NAME
Alabama	17.2	0.200000	0.012316
Alaska	7.2	0.200000	0.025758
Arizona	221.6	4.000000	0.018107
Arkansas	32.8	0.600000	0.009146
California	959.8	27.200000	0.028152
Colorado	134.6	2.300000	0.016700
Connecticut	131.5	3.800000	0.031444
Delaware	18.5	0.600000	0.030093
District of Columbia	97.9	1.000000	0.010958
Federal	14.0	0.000000	0.000000
Florida	118.7	1.400000	0.011173
Georgia	31.7	0.800000	0.027123
Guam	1.0	0.000000	0.000000
Hawaii	44.0	6.000000	0.136364
Idaho	32.1	0.400000	0.012256
Illinois	98.7	1.900000	0.017934
Indiana	75.3	1.700000	0.026156
Iowa	12.6	0.700000	0.055120
Kansas	74.2	1.400000	0.018572
Kentucky	204.2	2.900000	0.014771
Louisiana	22.3	0.100000	0.002500
Maine	40.2	0.500000	0.013078
Maryland	56.6	1.200000	0.021386
Massachusetts	362.7	12.000000	0.033657
Michigan	389.0	8.000000	0.021604
Minnesota	124.0	4.000000	0.030746
Mississippi	13.0	0.333333	0.048990
Missouri	104.0	1.400000	0.012809
Montana	28.8	0.000000	0.000000
Nebraska	39.2	1.000000	0.027177
Nevada	50.0	0.700000	0.015617
New Hampshire	21.7	1.100000	0.048537
New Jersey	451.8	9.500000	0.021666
New Mexico	19.1	0.400000	0.021558
New York	593.9	8.100000	0.013942
North Carolina	130.2	1.400000	0.010333
North Dakota	25.7	0.400000	0.011749
Ohio	346.5	3.700000	0.011713
Oklahoma	36.6	0.800000	0.017184
Oregon	106.3	1.100000	0.008745
Pennsylvania	53.2	1.900000	0.038109
Rhode Island	15.8	0.300000	0.018803
South Carolina	78.9	0.400000	0.005007
South Dakota	25.3	0.100000	0.004545
Tennessee	186.0	1.500000	0.008233
Texas	192.8	4.100000	0.021408
Utah	62.3	1.000000	0.015070
Vermont	20.0	0.400000	0.035049
Virginia	151.1	3.200000	0.022477
Washington	322.6	10.700000	0.033506

Select 2019 Anti Asian Rate in each state as our Observation:

df4 = df2.query('DATA_YEAR==2019').groupby(['STATE_NAME']).agg('mean')
df4 = df4.drop(columns=['BIAS_DESC', 'Anti-Asian-count'])
df4[:50]

	Anti-Asian-rate
STATE_NAME
Alabama	0.000000
Alaska	0.000000
Arizona	0.004785
Arkansas	0.029412
California	0.042157
Colorado	0.027650
Connecticut	0.023256
Delaware	0.000000
District of Columbia	0.004505
Federal	0.000000
Florida	0.017544
Georgia	0.028037
Hawaii	0.048780
Idaho	0.083333
Illinois	0.015385
Indiana	0.000000
Iowa	0.100000
Kansas	0.000000
Kentucky	0.013699
Louisiana	0.000000
Maine	0.000000
Maryland	0.000000
Massachusetts	0.033419
Michigan	0.013825
Minnesota	0.000000
Mississippi	0.111111
Missouri	0.010526
Montana	0.027027
Nebraska	0.000000
Nevada	0.045455
New Hampshire	0.062500
New Jersey	0.029382
New Mexico	0.000000
New York	0.009804
North Carolina	0.016736
North Dakota	0.055556
Ohio	0.017857
Oklahoma	0.000000
Oregon	0.039773
Pennsylvania	0.048780
Rhode Island	0.000000
South Carolina	0.000000
South Dakota	0.000000
Tennessee	0.000000
Texas	0.012605
Utah	0.000000
Vermont	0.000000
Virginia	0.023810
Washington	0.044199
West Virginia	0.018868

Combine them to data frame df4.

df4['Model_2009-2018']=df3['Anti-Asian-rate']
df4= df4.rename(columns={"Anti-Asian-rate": "Observed_2019"})
df4[:50]

	Observed_2019	Model_2009-2018
STATE_NAME
Alabama	0.000000	0.012316
Alaska	0.000000	0.025758
Arizona	0.004785	0.018107
Arkansas	0.029412	0.009146
California	0.042157	0.028152
Colorado	0.027650	0.016700
Connecticut	0.023256	0.031444
Delaware	0.000000	0.030093
District of Columbia	0.004505	0.010958
Federal	0.000000	0.000000
Florida	0.017544	0.011173
Georgia	0.028037	0.027123
Hawaii	0.048780	0.136364
Idaho	0.083333	0.012256
Illinois	0.015385	0.017934
Indiana	0.000000	0.026156
Iowa	0.100000	0.055120
Kansas	0.000000	0.018572
Kentucky	0.013699	0.014771
Louisiana	0.000000	0.002500
Maine	0.000000	0.013078
Maryland	0.000000	0.021386
Massachusetts	0.033419	0.033657
Michigan	0.013825	0.021604
Minnesota	0.000000	0.030746
Mississippi	0.111111	0.048990
Missouri	0.010526	0.012809
Montana	0.027027	0.000000
Nebraska	0.000000	0.027177
Nevada	0.045455	0.015617
New Hampshire	0.062500	0.048537
New Jersey	0.029382	0.021666
New Mexico	0.000000	0.021558
New York	0.009804	0.013942
North Carolina	0.016736	0.010333
North Dakota	0.055556	0.011749
Ohio	0.017857	0.011713
Oklahoma	0.000000	0.017184
Oregon	0.039773	0.008745
Pennsylvania	0.048780	0.038109
Rhode Island	0.000000	0.018803
South Carolina	0.000000	0.005007
South Dakota	0.000000	0.004545
Tennessee	0.000000	0.008233
Texas	0.012605	0.021408
Utah	0.000000	0.015070
Vermont	0.000000	0.035049
Virginia	0.023810	0.022477
Washington	0.044199	0.033506
West Virginia	0.018868	0.019253

Caculate P(O|M):

$P(O|M)\approx1-|O-E|$

df4[['Model_2009-2018','Observed_2019']] = df4[['Model_2009-2018','Observed_2019']]
df4['P(O|M)']=1-abs(df4['Model_2009-2018']-df4['Observed_2019'])
df4[:50]

	Observed_2019	Model_2009-2018	P(O|M)
STATE_NAME
Alabama	0.000000	0.012316	0.987684
Alaska	0.000000	0.025758	0.974242
Arizona	0.004785	0.018107	0.986677
Arkansas	0.029412	0.009146	0.979734
California	0.042157	0.028152	0.985995
Colorado	0.027650	0.016700	0.989050
Connecticut	0.023256	0.031444	0.991812
Delaware	0.000000	0.030093	0.969907
District of Columbia	0.004505	0.010958	0.993546
Federal	0.000000	0.000000	1.000000
Florida	0.017544	0.011173	0.993629
Georgia	0.028037	0.027123	0.999086
Hawaii	0.048780	0.136364	0.912417
Idaho	0.083333	0.012256	0.928923
Illinois	0.015385	0.017934	0.997451
Indiana	0.000000	0.026156	0.973844
Iowa	0.100000	0.055120	0.955120
Kansas	0.000000	0.018572	0.981428
Kentucky	0.013699	0.014771	0.998927
Louisiana	0.000000	0.002500	0.997500
Maine	0.000000	0.013078	0.986922
Maryland	0.000000	0.021386	0.978614
Massachusetts	0.033419	0.033657	0.999763
Michigan	0.013825	0.021604	0.992220
Minnesota	0.000000	0.030746	0.969254
Mississippi	0.111111	0.048990	0.937879
Missouri	0.010526	0.012809	0.997718
Montana	0.027027	0.000000	0.972973
Nebraska	0.000000	0.027177	0.972823
Nevada	0.045455	0.015617	0.970162
New Hampshire	0.062500	0.048537	0.986037
New Jersey	0.029382	0.021666	0.992284
New Mexico	0.000000	0.021558	0.978442
New York	0.009804	0.013942	0.995861
North Carolina	0.016736	0.010333	0.993596
North Dakota	0.055556	0.011749	0.956194
Ohio	0.017857	0.011713	0.993855
Oklahoma	0.000000	0.017184	0.982816
Oregon	0.039773	0.008745	0.968973
Pennsylvania	0.048780	0.038109	0.989329
Rhode Island	0.000000	0.018803	0.981197
South Carolina	0.000000	0.005007	0.994993
South Dakota	0.000000	0.004545	0.995455
Tennessee	0.000000	0.008233	0.991767
Texas	0.012605	0.021408	0.991197
Utah	0.000000	0.015070	0.984930
Vermont	0.000000	0.035049	0.964951
Virginia	0.023810	0.022477	0.998668
Washington	0.044199	0.033506	0.989307
West Virginia	0.018868	0.019253	0.999614

Caculate P(M|O):

$p(θ|y)=p(y|θ)p(θ)$

df4['P(M|O)']=df4['P(O|M)']*df4['Model_2009-2018']             
df4[:50]

	Observed_2019	Model_2009-2018	P(O|M)	P(M|O)
STATE_NAME
Alabama	0.000000	0.012316	0.987684	0.012164
Alaska	0.000000	0.025758	0.974242	0.025094
Arizona	0.004785	0.018107	0.986677	0.017866
Arkansas	0.029412	0.009146	0.979734	0.008960
California	0.042157	0.028152	0.985995	0.027758
Colorado	0.027650	0.016700	0.989050	0.016517
Connecticut	0.023256	0.031444	0.991812	0.031186
Delaware	0.000000	0.030093	0.969907	0.029188
District of Columbia	0.004505	0.010958	0.993546	0.010887
Federal	0.000000	0.000000	1.000000	0.000000
Florida	0.017544	0.011173	0.993629	0.011101
Georgia	0.028037	0.027123	0.999086	0.027098
Hawaii	0.048780	0.136364	0.912417	0.124420
Idaho	0.083333	0.012256	0.928923	0.011385
Illinois	0.015385	0.017934	0.997451	0.017888
Indiana	0.000000	0.026156	0.973844	0.025472
Iowa	0.100000	0.055120	0.955120	0.052647
Kansas	0.000000	0.018572	0.981428	0.018227
Kentucky	0.013699	0.014771	0.998927	0.014755
Louisiana	0.000000	0.002500	0.997500	0.002494
Maine	0.000000	0.013078	0.986922	0.012907
Maryland	0.000000	0.021386	0.978614	0.020928
Massachusetts	0.033419	0.033657	0.999763	0.033649
Michigan	0.013825	0.021604	0.992220	0.021436
Minnesota	0.000000	0.030746	0.969254	0.029800
Mississippi	0.111111	0.048990	0.937879	0.045947
Missouri	0.010526	0.012809	0.997718	0.012779
Montana	0.027027	0.000000	0.972973	0.000000
Nebraska	0.000000	0.027177	0.972823	0.026438
Nevada	0.045455	0.015617	0.970162	0.015151
New Hampshire	0.062500	0.048537	0.986037	0.047859
New Jersey	0.029382	0.021666	0.992284	0.021499
New Mexico	0.000000	0.021558	0.978442	0.021093
New York	0.009804	0.013942	0.995861	0.013885
North Carolina	0.016736	0.010333	0.993596	0.010266
North Dakota	0.055556	0.011749	0.956194	0.011234
Ohio	0.017857	0.011713	0.993855	0.011641
Oklahoma	0.000000	0.017184	0.982816	0.016889
Oregon	0.039773	0.008745	0.968973	0.008474
Pennsylvania	0.048780	0.038109	0.989329	0.037702
Rhode Island	0.000000	0.018803	0.981197	0.018450
South Carolina	0.000000	0.005007	0.994993	0.004982
South Dakota	0.000000	0.004545	0.995455	0.004525
Tennessee	0.000000	0.008233	0.991767	0.008166
Texas	0.012605	0.021408	0.991197	0.021220
Utah	0.000000	0.015070	0.984930	0.014843
Vermont	0.000000	0.035049	0.964951	0.033821
Virginia	0.023810	0.022477	0.998668	0.022447
Washington	0.044199	0.033506	0.989307	0.033148
West Virginia	0.018868	0.019253	0.999614	0.019246

Caculate $Surprise=D_{KL}(P(M|O)||P(M))$:

from scipy.stats import entropy
import numpy as np
df4['Signed_Surprise']= df4['P(M|O)']*np.log(df4['P(M|O)']/df4['Model_2009-2018'])*np.sign(df4['Model_2009-2018']-df4['Observed_2019'])
df4[:80]

	Observed_2019	Model_2009-2018	P(O|M)	P(M|O)	Signed_Surprise
STATE_NAME
Alabama	0.000000	0.012316	0.987684	0.012164	-0.000151
Alaska	0.000000	0.025758	0.974242	0.025094	-0.000655
Arizona	0.004785	0.018107	0.986677	0.017866	-0.000240
Arkansas	0.029412	0.009146	0.979734	0.008960	0.000183
California	0.042157	0.028152	0.985995	0.027758	0.000391
Colorado	0.027650	0.016700	0.989050	0.016517	0.000182
Connecticut	0.023256	0.031444	0.991812	0.031186	-0.000256
Delaware	0.000000	0.030093	0.969907	0.029188	-0.000892
District of Columbia	0.004505	0.010958	0.993546	0.010887	-0.000070
Federal	0.000000	0.000000	1.000000	0.000000	NaN
Florida	0.017544	0.011173	0.993629	0.011101	0.000071
Georgia	0.028037	0.027123	0.999086	0.027098	0.000025
Hawaii	0.048780	0.136364	0.912417	0.124420	-0.011404
Idaho	0.083333	0.012256	0.928923	0.011385	0.000839
Illinois	0.015385	0.017934	0.997451	0.017888	-0.000046
Indiana	0.000000	0.026156	0.973844	0.025472	-0.000675
Iowa	0.100000	0.055120	0.955120	0.052647	0.002417
Kansas	0.000000	0.018572	0.981428	0.018227	-0.000342
Kentucky	0.013699	0.014771	0.998927	0.014755	-0.000016
Louisiana	0.000000	0.002500	0.997500	0.002494	-0.000006
Maine	0.000000	0.013078	0.986922	0.012907	-0.000170
Maryland	0.000000	0.021386	0.978614	0.020928	-0.000452
Massachusetts	0.033419	0.033657	0.999763	0.033649	-0.000008
Michigan	0.013825	0.021604	0.992220	0.021436	-0.000167
Minnesota	0.000000	0.030746	0.969254	0.029800	-0.000931
Mississippi	0.111111	0.048990	0.937879	0.045947	0.002947
Missouri	0.010526	0.012809	0.997718	0.012779	-0.000029
Montana	0.027027	0.000000	0.972973	0.000000	NaN
Nebraska	0.000000	0.027177	0.972823	0.026438	-0.000728
Nevada	0.045455	0.015617	0.970162	0.015151	0.000459
New Hampshire	0.062500	0.048537	0.986037	0.047859	0.000673
New Jersey	0.029382	0.021666	0.992284	0.021499	0.000167
New Mexico	0.000000	0.021558	0.978442	0.021093	-0.000460
New York	0.009804	0.013942	0.995861	0.013885	-0.000058
North Carolina	0.016736	0.010333	0.993596	0.010266	0.000066
North Dakota	0.055556	0.011749	0.956194	0.011234	0.000503
Ohio	0.017857	0.011713	0.993855	0.011641	0.000072
Oklahoma	0.000000	0.017184	0.982816	0.016889	-0.000293
Oregon	0.039773	0.008745	0.968973	0.008474	0.000267
Pennsylvania	0.048780	0.038109	0.989329	0.037702	0.000404
Rhode Island	0.000000	0.018803	0.981197	0.018450	-0.000350
South Carolina	0.000000	0.005007	0.994993	0.004982	-0.000025
South Dakota	0.000000	0.004545	0.995455	0.004525	-0.000021
Tennessee	0.000000	0.008233	0.991767	0.008166	-0.000068
Texas	0.012605	0.021408	0.991197	0.021220	-0.000188
Utah	0.000000	0.015070	0.984930	0.014843	-0.000225
Vermont	0.000000	0.035049	0.964951	0.033821	-0.001207
Virginia	0.023810	0.022477	0.998668	0.022447	0.000030
Washington	0.044199	0.033506	0.989307	0.033148	0.000356
West Virginia	0.018868	0.019253	0.999614	0.019246	-0.000007
Wisconsin	0.000000	0.027947	0.972053	0.027166	-0.000770
Wyoming	0.000000	0.000000	1.000000	0.000000	NaN

df4['Signed_Surprise'] = df4['Signed_Surprise'].fillna(0)
df4['State_Name'] = df4.index
df4[:80]

	Observed_2019	Model_2009-2018	P(O|M)	P(M|O)	Signed_Surprise	State_Name
STATE_NAME
Alabama	0.000000	0.012316	0.987684	0.012164	-0.000151	Alabama
Alaska	0.000000	0.025758	0.974242	0.025094	-0.000655	Alaska
Arizona	0.004785	0.018107	0.986677	0.017866	-0.000240	Arizona
Arkansas	0.029412	0.009146	0.979734	0.008960	0.000183	Arkansas
California	0.042157	0.028152	0.985995	0.027758	0.000391	California
Colorado	0.027650	0.016700	0.989050	0.016517	0.000182	Colorado
Connecticut	0.023256	0.031444	0.991812	0.031186	-0.000256	Connecticut
Delaware	0.000000	0.030093	0.969907	0.029188	-0.000892	Delaware
District of Columbia	0.004505	0.010958	0.993546	0.010887	-0.000070	District of Columbia
Federal	0.000000	0.000000	1.000000	0.000000	0.000000	Federal
Florida	0.017544	0.011173	0.993629	0.011101	0.000071	Florida
Georgia	0.028037	0.027123	0.999086	0.027098	0.000025	Georgia
Hawaii	0.048780	0.136364	0.912417	0.124420	-0.011404	Hawaii
Idaho	0.083333	0.012256	0.928923	0.011385	0.000839	Idaho
Illinois	0.015385	0.017934	0.997451	0.017888	-0.000046	Illinois
Indiana	0.000000	0.026156	0.973844	0.025472	-0.000675	Indiana
Iowa	0.100000	0.055120	0.955120	0.052647	0.002417	Iowa
Kansas	0.000000	0.018572	0.981428	0.018227	-0.000342	Kansas
Kentucky	0.013699	0.014771	0.998927	0.014755	-0.000016	Kentucky
Louisiana	0.000000	0.002500	0.997500	0.002494	-0.000006	Louisiana
Maine	0.000000	0.013078	0.986922	0.012907	-0.000170	Maine
Maryland	0.000000	0.021386	0.978614	0.020928	-0.000452	Maryland
Massachusetts	0.033419	0.033657	0.999763	0.033649	-0.000008	Massachusetts
Michigan	0.013825	0.021604	0.992220	0.021436	-0.000167	Michigan
Minnesota	0.000000	0.030746	0.969254	0.029800	-0.000931	Minnesota
Mississippi	0.111111	0.048990	0.937879	0.045947	0.002947	Mississippi
Missouri	0.010526	0.012809	0.997718	0.012779	-0.000029	Missouri
Montana	0.027027	0.000000	0.972973	0.000000	0.000000	Montana
Nebraska	0.000000	0.027177	0.972823	0.026438	-0.000728	Nebraska
Nevada	0.045455	0.015617	0.970162	0.015151	0.000459	Nevada
New Hampshire	0.062500	0.048537	0.986037	0.047859	0.000673	New Hampshire
New Jersey	0.029382	0.021666	0.992284	0.021499	0.000167	New Jersey
New Mexico	0.000000	0.021558	0.978442	0.021093	-0.000460	New Mexico
New York	0.009804	0.013942	0.995861	0.013885	-0.000058	New York
North Carolina	0.016736	0.010333	0.993596	0.010266	0.000066	North Carolina
North Dakota	0.055556	0.011749	0.956194	0.011234	0.000503	North Dakota
Ohio	0.017857	0.011713	0.993855	0.011641	0.000072	Ohio
Oklahoma	0.000000	0.017184	0.982816	0.016889	-0.000293	Oklahoma
Oregon	0.039773	0.008745	0.968973	0.008474	0.000267	Oregon
Pennsylvania	0.048780	0.038109	0.989329	0.037702	0.000404	Pennsylvania
Rhode Island	0.000000	0.018803	0.981197	0.018450	-0.000350	Rhode Island
South Carolina	0.000000	0.005007	0.994993	0.004982	-0.000025	South Carolina
South Dakota	0.000000	0.004545	0.995455	0.004525	-0.000021	South Dakota
Tennessee	0.000000	0.008233	0.991767	0.008166	-0.000068	Tennessee
Texas	0.012605	0.021408	0.991197	0.021220	-0.000188	Texas
Utah	0.000000	0.015070	0.984930	0.014843	-0.000225	Utah
Vermont	0.000000	0.035049	0.964951	0.033821	-0.001207	Vermont
Virginia	0.023810	0.022477	0.998668	0.022447	0.000030	Virginia
Washington	0.044199	0.033506	0.989307	0.033148	0.000356	Washington
West Virginia	0.018868	0.019253	0.999614	0.019246	-0.000007	West Virginia
Wisconsin	0.000000	0.027947	0.972053	0.027166	-0.000770	Wisconsin
Wyoming	0.000000	0.000000	1.000000	0.000000	0.000000	Wyoming

USA 2019 Asian Hate Crimes Rate Observed Heat Map

import plotly.express as px

fig0 = px.choropleth(df4, geojson=states, locations='State_Name', color='Observed_2019',featureidkey="properties.name",
                           color_continuous_scale="blues",
                           scope="usa",
                           labels={"USA 2019 Asian Hate Crimes Rate Observed Heat Map"}
                          )
fig0.update_layout(margin={"r":0,"t":0,"l":0,"b":0})

USA 2019 Asian Hate Crimes Bayesian Surprise Map based on last 10 years

fig1 = px.choropleth(df4, geojson=states, locations='State_Name', color='Signed_Surprise',featureidkey="properties.name",
                           color_continuous_scale="rdbu",
                           range_color=(-0.003,0.003),
                           scope="usa",
                           labels={"USA 2019 Asian Hate Crimes Bayesian Surprise Map based on last 10 years"}
                          )
fig1.update_layout(margin={"r":0,"t":0,"l":0,"b":0})

Different obejctive: USA 2019 Asian Hate Crimes Bayesian Surprise Map based on last 20 years

df3 = df2.query('DATA_YEAR>=1999 & DATA_YEAR<=2018').groupby(['STATE_NAME']).agg('mean')

df5 = df2.query('DATA_YEAR==2019').groupby(['STATE_NAME']).agg('mean')
df5 = df5.drop(columns=['BIAS_DESC', 'Anti-Asian-count'])

df5['Model_1999-2018']=df3['Anti-Asian-rate']
df5= df5.rename(columns={"Anti-Asian-rate": "Observed_2019"})

df5[['Model_1999-2018','Observed_2019']] = df5[['Model_1999-2018','Observed_2019']]
df5['P(O|M)']=1-abs(df5['Model_1999-2018']-df5['Observed_2019'])

df5['P(M|O)']=df5['P(O|M)']*df5['Model_1999-2018']             

df5['Signed_Surprise']= df5['P(M|O)']*np.log(df5['P(M|O)']/df5['Model_1999-2018'])*np.sign(df5['Model_1999-2018']-df5['Observed_2019'])

df5['Signed_Surprise'] = df5['Signed_Surprise'].fillna(0)
df5['State_Name'] = df5.index

df5[:80]

fig2 = px.choropleth(df5, geojson=states, locations='State_Name', color='Signed_Surprise',featureidkey="properties.name",
                           color_continuous_scale="rdbu",
                           range_color=(-0.003,0.003),
                           scope="usa",
                           labels={"2019 Asian Hate Crimes Bayesian Surprise Map based on last 20 years"}
                          )
fig2.update_layout(margin={"r":0,"t":0,"l":0,"b":0})

Comparing 3 maps

fig0.show() 
fig1.show()
fig2.show()